GSM-noise "buzz" issue / ways forward

Andy Green andy at
Mon Sep 22 11:35:06 CEST 2008

Hash: SHA1

Somebody in the thread at some point said:

| Only answer I'm interested in is: why don't we already have those
| beads/EMI-filters in GTA02A7, though it's been validated to fix buzz

What was actually done on A7 revision?  For example, what about the
increased caps for Vref Vmid, are they present on A7 (or A6 for that
matter)?  The BOM teenie put on internal Wiki seems to be gone.

I heard that there was some reduction in audio noise on A7, but what was
that exactly?

Why don't we do more of the communications taking place about these
things in public.  That way there can be better understanding not only
internally (since private mails on this do not reach me either) but
evidently we have people with real RF expertise on this list and there
is every reason to let them know the details we find.

| more than a month now, and we seemingly all accepted it's a viable fix
| GSM-buzz (or at least we agree it's safe to have those beads even for
| U4906, U4907!)? When are we going to give them a try on GTA02? Is anybody
| assuming we could spoil GTA02 design to a point where even replacing
| EMI-filters by 0R would break the device?

Adding the bead is one of only two things that seemed to impact the
symptom, the other being removal of headset jack.

Has the same bead been tried on the headset mic net and did it help there?

| I urgently recommend to start to do a PCB with pads for EMI-filters *now*.
| I absolutely veto on producing GTA02 with a 100uF on C4306.

For those reading along at home, C4306 is new on GTA02 A7 and is between
MIC1 net and 0V.

I agree it's not clear 100uF there is going to do anything good about
the problem, unless the dips are introduced directly on MICBIAS then
trying to smooth them there is the wrong place since they have already
been amplified on to there and we fight the blameless amplifier.  But
actually, I didn't see we proved quite where the buzz starts from as LF
so we are not in a position to write this off as insane.

I found I can reliably make buzz by scope probe on Vref, which is the
source for input of MICBIAS amp, that's a bit of a clue I think.

| @Sean: you asked me to solve this issue. That's my result for now.

For going forward, we did reportedly find the bead helped for internal
mic.  So Joerg's point is right if we are looking at A8 then you would
make a place for a bead on the PCB same as done for GTA03.  It's the
only doable thing that actually impacted anything we saw so far.  I
don't know it helps for external mic net, or if it is really a "fix",
but it should do something for half the symptoms anyway.

For both going forward and to solve issue for phones in the field, I get
the feeling we can't do justice to understanding this problem internally
and need to go talk to people experienced in this kind of RF issue.  Uwe
and his folks sound like a first port of call.  I know a commercial
outfit in UK I worked with before we can approach as well if needed.

- -Andy
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora -


More information about the hardware mailing list